I can get a 1000 MH ballast for under $50 CAD. I hear alot of mixed feelings about using such a high wattage with hotspotting and uneven ligt distribution, but others are saying they are fine. Does anyone have any input on this, maybe some personal experience. Hopefully i get replies soon because i have to let the guy know before tomorow.
1000 watt Metal alide setup
- Thread starter juice_e
- Start date
This is speculation on my part. Good luck with it.
I'm thinking about the size of the colour pixels and how much light they can pass. There is probably a flux density that is optimal. More may just heat the lcd.
Anyone compared a 7" with 250w vs 7" with 400w vs 150w? Do you actually get much more light out?
I'm thinking about the size of the colour pixels and how much light they can pass. There is probably a flux density that is optimal. More may just heat the lcd.
Anyone compared a 7" with 250w vs 7" with 400w vs 150w? Do you actually get much more light out?
would a 250 watt electronic ballast with a double ended HQI bulb give off alot less heat than a 250 watt coil & core ballast with an ordinary big bulb? If i can get coil,core ballast for $50 or HQI ballast for $150, is it worth the difference? And will the small double ended bulb be too small or not enough light for a 15" LCD. T
his is very important question for me because it is the last step im waiting for until i start building. Im stuff in a bind here.
his is very important question for me because it is the last step im waiting for until i start building. Im stuff in a bind here.
more light? less light?
Determine the lamp to use by the size image you want. Lumens per square foot drops by the square of the diagonal size. So a 60" diagonal image will be four times brighter than a 120" image.
If you will have a big image, then get a powerful lamp. I doubt there is any sort of "step function" in LCD light throughput. With more light, the output will go up by the same amount. So will the generation of heat within the LCD. Make sure you have a good IR filter in your projector. Same goes for UV.
The size of the LCD will have an effect on the energy absorbed per square inch. A 15" LCD has more than 4 times the surface area of a 7" LCD. With a small LCD and a big lamp, you may need a fan blowing air right over the LCD surface.
A good electronic ballast will give you a bit more light than a magnetic core ballast. It runs the lamp closer to the edge, because it regulates the power much better. Neither type of ballast will put out a lot of heat. Nothing like the lamp itself.
Determine the lamp to use by the size image you want. Lumens per square foot drops by the square of the diagonal size. So a 60" diagonal image will be four times brighter than a 120" image.
If you will have a big image, then get a powerful lamp. I doubt there is any sort of "step function" in LCD light throughput. With more light, the output will go up by the same amount. So will the generation of heat within the LCD. Make sure you have a good IR filter in your projector. Same goes for UV.
The size of the LCD will have an effect on the energy absorbed per square inch. A 15" LCD has more than 4 times the surface area of a 7" LCD. With a small LCD and a big lamp, you may need a fan blowing air right over the LCD surface.
A good electronic ballast will give you a bit more light than a magnetic core ballast. It runs the lamp closer to the edge, because it regulates the power much better. Neither type of ballast will put out a lot of heat. Nothing like the lamp itself.
There is controversy over how dynamic/static a measure contrast is, and whether that would cause overexposure.
But aside from that, there's the issue of the size of the arc. Most industrial or hydroponic lighting has long arcs and corresponding long life 1000 watt lamps. 80mm or more is typical. While our triplets can focus a 25-40mm arc which 400w non-HQI lamps present it with, I don't think that the long arc of a longlife 1000w lamp will be effective.
I tracked down the 31mm arc OSRAM HQI-TS d/s lamp a while back, but eventually concluded that it would be too expensive. It's a stadium/spotlight lamp w/ a 6000 hour or so lifetime, good color rendering, the works. OSRAM's UK site mentions that it's the lamp used for the last Olympic games. But it's rather expensive - I concluded that I couldn't afford the $400 or so it would have come out to for lamp, ballast, and igniter, plus $100 for a 120v->240v transformer (only ballast it has runs on 240v). This was after I found a decently priced place to buy the lamp, the MSRP is apparently $700 something just for the lamp, and data was sketchy. If you want details, send me a PM.
But aside from that, there's the issue of the size of the arc. Most industrial or hydroponic lighting has long arcs and corresponding long life 1000 watt lamps. 80mm or more is typical. While our triplets can focus a 25-40mm arc which 400w non-HQI lamps present it with, I don't think that the long arc of a longlife 1000w lamp will be effective.
I tracked down the 31mm arc OSRAM HQI-TS d/s lamp a while back, but eventually concluded that it would be too expensive. It's a stadium/spotlight lamp w/ a 6000 hour or so lifetime, good color rendering, the works. OSRAM's UK site mentions that it's the lamp used for the last Olympic games. But it's rather expensive - I concluded that I couldn't afford the $400 or so it would have come out to for lamp, ballast, and igniter, plus $100 for a 120v->240v transformer (only ballast it has runs on 240v). This was after I found a decently priced place to buy the lamp, the MSRP is apparently $700 something just for the lamp, and data was sketchy. If you want details, send me a PM.
I doubt anyone expects a step function. I dont expect a linear fuction either. As the heat level goes up the lcd will be affected at the level of the twisting molecules and viscosity of the substrate. The lcd manufactures would know the impact but i havent found any tech specs published on the net. If someone has found this info please let everyone know.
If someone can measure output with 150, 250, 400w through the same setup you can curve fit to those 3 points and extrapolate to 1000w.
increasing output
The limitation with the DIY designs, is that we use LCD panels that have integrated color filters and polarized filters. So any light that doesn't get through the image generator (the LCD panel) gets absorbed right at the LCD cells.
Higher-powered commercial projectors use LCDs without color filters and without polarized filters. Seperate color filters pass red, green, or blue light to one of three LCD arrays. All the LCD cells do is twist the polarity of the light to some degree. (Which absorbs very little energy.) Then a seperate heat-resistant polarized filter passes or blocks the light.
This approach is not beyond DIY: The interference filters, beam splitter & combiner, etc. might be too difficult. We could trade off some design complexity to make it easier to build, but less efficient. For example, use three 150 Watt MH lamps, so each LCD has its own light source. Use three of those 5" diameter CRT projection lenses, so the light paths can just be combined on the screen. The tricky part would be building or adapting an LCD controller to drive each panel with only one color's signals.
The limitation with the DIY designs, is that we use LCD panels that have integrated color filters and polarized filters. So any light that doesn't get through the image generator (the LCD panel) gets absorbed right at the LCD cells.
Higher-powered commercial projectors use LCDs without color filters and without polarized filters. Seperate color filters pass red, green, or blue light to one of three LCD arrays. All the LCD cells do is twist the polarity of the light to some degree. (Which absorbs very little energy.) Then a seperate heat-resistant polarized filter passes or blocks the light.
This approach is not beyond DIY: The interference filters, beam splitter & combiner, etc. might be too difficult. We could trade off some design complexity to make it easier to build, but less efficient. For example, use three 150 Watt MH lamps, so each LCD has its own light source. Use three of those 5" diameter CRT projection lenses, so the light paths can just be combined on the screen. The tricky part would be building or adapting an LCD controller to drive each panel with only one color's signals.
wouldnt work
You can go 3 seperate systems with colour lcds or 1 lamp with light split to 1 colour lcds. You can't mix them.
Overlaping 3 systems would be a nightmare for aligning everything. Every part of each has to be perfecto or you get ghosting.
You're more likely to get 3 systems side by side and be able to crop the edges.
You can go 3 seperate systems with colour lcds or 1 lamp with light split to 1 colour lcds. You can't mix them.
Overlaping 3 systems would be a nightmare for aligning everything. Every part of each has to be perfecto or you get ghosting.
You're more likely to get 3 systems side by side and be able to crop the edges.
I myself am thinking about going the "grow lamp" route, but the bulb is 11.5" long! Right now I am "testing" with a 500 watt halogen and that is putting out about 12000 lumens, but way too much heat. Has anyone had good success with the larger type bulb I am thinking aobut? I mean, I can get the whole kit for $160 plus shipping to my door, so for that price and the fact that it will be about 3 time more lumen (so lets say twice as bright final result) is it going to be worth the hassle of trying to get a huge bulb like that to work? If it has been done, I'dlike to see some dteailed pics of the relfector used in the set up too...
Right now I am using a stanless steel bowl from a dollar store and th guts of the 500 watt work light are screwed into that, made the picture much brighter, but there is a slight dark line diagonally (where the bulbisin the reflector...and that is something I'd like to avoid if I get an even larger bulb (albiet with a smaller arc).
Right now I am using a stanless steel bowl from a dollar store and th guts of the 500 watt work light are screwed into that, made the picture much brighter, but there is a slight dark line diagonally (where the bulbisin the reflector...and that is something I'd like to avoid if I get an even larger bulb (albiet with a smaller arc).